A joint optimization method for UAV-assisted data collection and trajectory planning in wireless networks
By optimizing the flight trajectory of the drone in drone-assisted data collection and trajectory planning, the problem of drone effectively completing data acquisition tasks within a limited communication range is solved, and the effect of reducing energy consumption and improving energy efficiency is achieved.
Patent Information
- Application Number
- CN202110073450.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-01-20
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2041-01-20
AI Technical Summary
When performing data collection tasks, the communication link is limited, and each ground user's effective communication range is limited, resulting in the drone being able to transmit data with the ground user only when overflowing. How to effectively complete it within a limited communication range Data acquisition tasks have become the main technical problem.
Using a joint optimization method of data collection and trajectory planning based on drone assistance, the optimal waypoint and projection method are designed by optimizing the flight trajectory of the drone, using the three-dimensional environment and the communication range model of the ground users, the waypoint and projection method are designed to calculate the optimal waypoint To improve data transmission efficiency.
By optimizing the flight trajectory, energy consumption during drone communication is reduced, energy efficiency is improved, and data acquisition tasks are effectively completed.
Smart Images

Figure CN114859957B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of trajectory optimization when using a drone to collect data in a drone-assisted wireless network, and is mainly used for planning an optimal flight trajectory for a drone while ensuring fair transmission for all ground users when the drone performs a data collection task. Background Art
[0002] Unmanned aerial vehicles (UAVs) have been developing and penetrating into various fields in the past few years, such as cargo transportation, video transmission, detection and surveillance, search and rescue, etc. As it can be equipped with communication devices and large-capacity batteries, UAVs are becoming more and more popular in wireless networks to meet the quality of service (QoS) requirements of on-demand deployment. Based on these advantages, UAVs can act as data collection carriers flying over dispersed ground users (GUs), which provide short-range line-of-sight (LoS) communication links and ensure the completion of assigned data collection tasks. However, due to the limitations of the communication link, such as distance and minimum required signal-to-noise ratio (SNR), the effective communication range of each GU is limited, and UAVs can only transmit data with GUs when flying over their communication range. Therefore, how to effectively complete the data collection task within a limited communication range through UAV trajectory optimization has attracted the attention of many researchers. Summary of the invention
[0003] Purpose of the invention: Due to the limitation of communication links, the effective communication range of each ground user is limited, and the UAV can transmit data with the ground user only when it flies over its communication range. Therefore, how to effectively complete the data collection task within the limited communication range through UAV trajectory optimization has become a major technical issue.
[0004] To solve this technical problem, the present invention proposes a joint optimization method for data collection and trajectory planning based on drone assistance in a wireless network, which effectively completes the data collection task by optimizing the trajectory.
[0005] Technical solution: To achieve the above technical effects, the technical solution proposed by the present invention is:
[0006] A method for joint optimization of data collection and trajectory planning based on drone assistance in wireless networks, comprising the steps of:
[0007] (1) Construct the system model (assuming that the UAV flies in a three-dimensional environment, the initial position of the UAV is at the origin (0, 0, 0), and the final position is (x F ,y F , H), further, assuming that the UAV flies horizontally at a fixed height H and a constant speed V, N ground users collect data represented by N = {1, ..., N}, and the communication range of the nth ground user is represented by a radius of r n Disc Dn ):
[0008] 1) The time-varying distance from the nth ground user to the UAV is:
[0009]
[0010] Where u(t) = [x(t), y(t)] T represents the coordinates of the above position projected on the horizontal plane, g n =[a n , b n ] T Indicates the position of the nth terrestrial user.
[0011] 2) The instantaneous channel power gain from the nth ground user to the UAV is:
[0012]
[0013] Wherein, β0 represents the channel power gain when the reference distance d0=1m. B represents the total available channel bandwidth in Hertz (Hz).
[0014] 3) The instantaneous channel capacity from the UAV to the nth ground user is:
[0015]
[0016] where P is the transmit power from the ground user to the UAV, which is assumed to be constant during mission completion, and σ 2 is the additive white Gaussian noise (AWGN) of the nth terrestrial user uplink, γ0=β0P / σ 2 is the reference received SNR at d0=1m.
[0017] 4) The constraints are:
[0018]
[0019] Among them, the binary variable ξ n represents the communication execution period between the UAV and each ground user during the mission. If the UAV flies beyond the range D n If the value is the inner area of , it is equal to 1, otherwise it is equal to 0.
[0020] 5) The total amount of data sent from the nth drone to the drone within the duration T is:
[0021]
[0022] (2) Waypoint positioning based on parallel projection. We transform the trajectory design problem into the location problem of a set of waypoints, which are connected with straight lines in sequence to form the trajectory of the UAV. We will adjust the parallel projection method based on the ground topology to calculate the location of the waypoints by reformulating some of the parallel projection method rules.
[0023] 1) The starting point x is in the convex set D n The only orthogonal projection onto is:
[0024]
[0025] 2) The single-step parallel projection function is:
[0026]
[0027] Among them, β l is the iteration point of PPM, λ l ∈[ε, 2-ε] where 0<ε<1 is called the relaxation parameter. Geometrically, it represents β l Move to Beta l+1 The parallel projection method is an improvement on the projection method in formula (6), which projects to all disks D in each iteration. n The centroid of the polygon formed by these N projected points is used as the next iteration point.
[0028] 3) The improved parallel projection method is:
[0029]
[0030] Here, we assume that L = {1, ..., L} is the waypoint β on the trajectory of the UAV. l and β 1 =u0,β L =u F , β l =[x l ,y l ] T When input D n It will be used for each D in each MPPM iteration. n Projection to g n position, but not exceeding g n . Leave D n After that, we only need to F In other cases, the current waypoint β l The closest D to it is projected in the traditional projection method n at the boundary point.
[0031] 4) N projection points and waypoints β l+1The min-max problem between is:
[0032]
[0033] Where β = [x, y] T is the optimization variable.
[0034] 5) β l The iterative equation is:
[0035]
[0036] From formula (9), we can get the optimal solution β * After that, we still use λ l To control β l The step length β * .
[0037] (3) Two trajectory design methods are designed for homogeneous and heterogeneous ground topologies respectively:
[0038] For a homogeneous ground topology, each user has the same communication radius r n = r and the amount of data collection is unknown. In this case, we only need to directly use the modified parallel projection method to calculate the waypoints of the trajectory.
[0039] 1) Define n and l as the number of loops (the number of ground users and the number of iterations), and the ground user position g of each n , communication radius r n and the amount of data R n ; Define the initial position u0 and final position u of the drone F ;
[0040] 2) Define the step size λ, algorithm accuracy ε and maximum number of iterations I; initialize the starting waypoint β 1 =u0; initialize l = 2;
[0041] 3) for 1<n<N, calculate D according to formula (8) in 1 n n projection points on
[0042] 4) By solving the min-max problem in formula (9), we obtain the l+1th optimal solution β * :
[0043] β l+1 =β l +λ l (β * -β l ),
[0044] 5) l = l + 1, if l < I, return to step 3), otherwise, execute step 6);
[0045] 6) Return the waypoint set L.
[0046] For heterogeneous ground topology, each ground user may have a different communication radius r n and the preset amount of data collection. Therefore, the drone adds a new communication constraint during flight, namely, throughput, which can be expressed as:
[0047]
[0048] Where R n Indicates the amount of data for the nth item.
[0049] To ensure communication fairness under throughput constraints, N projection points and waypoints β l+1 The min-max problem between can be refactored as:
[0050]
[0051] where ω n When the communication radius r is given n and the amount of data R n This makes the waypoints closer and the waypoints have a larger data volume R n and a smaller communication radius r n According to the characteristics of the min-max problem, the larger the distance, the more adjustments are made to the direction. Therefore, ω n It is expressed as:
[0052]
[0053] For heterogeneous ground topology, the trajectory design method steps are as follows:
[0054] 1) Define n and l as the number of loops (the number of ground users and the number of iterations), and each ground user position g n , communication radius r n and the amount of data R n ; Define the initial position u0 and final position u of the drone F ;
[0055] 2) Define the step size λ, algorithm accuracy ε and maximum number of iterations I; initialize the starting waypoint β 1 =u0; initialize l = 2;
[0056] 3) for 1<n<N, calculate D according to formula (8) n n projection points on
[0057] 4) By solving the min-max problem in formula (12), the l+1th optimal solution β is obtained * :
[0058] 5) for 1<n<N, calculate the drone from β according to formula (5) l To β l+1 Total amount of data collected during the flight And calculate If R n <0, then delete g from the ground user N n ;
[0059] 6) l = l + 1, if l < I, return to step 3), otherwise, execute step 7);
[0060] 7) Return the waypoint set L.
[0061] Beneficial effects: The present invention can reduce the energy consumption during UAV communication and improve energy efficiency by optimizing the flight trajectory. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] Figure 1 Schematic diagram of a wireless network link between a drone and multiple ground users.
[0063] Figure 2 is a graphical illustration of the parallel projection algorithm.
[0064] Figure 3 It is a principle flow chart of the present invention based on a homogeneous ground topology structure.
[0065] Figure 4 It is a principle flow chart of the present invention based on heterogeneous ground topology.
[0066] Figure 5 It is a trajectory diagram of a drone collecting data from six ground users based on a homogeneous ground topology.
[0067] Figure 6 It is a trajectory diagram of a drone collecting data from six ground users based on a heterogeneous ground topology. DETAILED DESCRIPTION
[0068] The present invention will be further described below in conjunction with the accompanying drawings.
[0069] Figure 1 Schematic diagram of wireless network link between UAV and multiple ground users. Without loss of generality, we construct a 3-D rectangular coordinate system, where the predetermined initial position of the UAV is located at the origin (0, 0, 0), and assume that the UAV flies from the initial position to the determined final position (x(t), y(t), H) at a constant speed V and height H above the ground. We will denote (xF ,y F , H) is represented as the horizontal coordinate of the UAV changing with time in the total mission execution time T, where 0<t<T. In addition, the horizontal coordinate of the nth ground user is represented as (a n , b n , 0), and assume that the locations and communication radius of all ground users are known, and we assume that the final location and the entire ground topology are in the first quadrant of the two-dimensional coordinate.
[0070] Figure 2 The figure on the left shows the projection process from a starting point to all disks, while the blue triangles in the figure on the right are the result points of a single parallel projection. The parallel projection method is an improvement on the orthogonal projection method, which projects to all disks D in each iteration. n The centroid of the polygon formed by these N projected points is used as the next iteration point.
[0071] Figure 3 The figure is a flow chart of the principle of the present invention based on a homogeneous ground topology structure.
[0072] For a homogeneous ground topology, each user has the same communication radius r n = r and the amount of data collected is unknown. In this case, we only need to directly use the improved parallel projection method to calculate the waypoints of the trajectory. The specific steps are as follows:
[0073] 1) Define n and l as the number of loops (the number of ground users and the number of iterations), and the ground user position g of each n , communication radius r n and the amount of data R n ; Define the initial position u0 and final position u of the drone F ;
[0074] 2) Define the step size λ, algorithm accuracy ε and maximum number of iterations I; initialize the starting waypoint β 1 =u0; initialize l = 2;
[0075] 3) for 1<n<N, calculate D according to formula (8) in 1 n n projection points on
[0076] 4) By solving the min-max problem in formula (9), we obtain the l+1th optimal solution β * :
[0077] β l+1 =β l +λ l (β * -β l),
[0078] 5) l = l + 1, if l < I, return to step 3), otherwise, execute step 6);
[0079] 6) Return the waypoint set L.
[0080] Figure 4 The figure is a flow chart of the principle of the present invention based on heterogeneous ground topology.
[0081] For heterogeneous ground topology, each ground user may have a different communication radius r n and the preset amount of data collection. Therefore, the drone adds a new communication constraint during flight, namely, throughput, which can be expressed as:
[0082]
[0083] Where R n Indicates the amount of data for the nth item.
[0084] To ensure communication fairness under throughput constraints, N projection points and waypoints β l+1 The min-max problem between can be refactored as:
[0085]
[0086] st
[0087]
[0088] where ω n When the communication radius r is given n and the amount of data R n This makes the waypoints closer and the waypoints have a larger data volume R n and a smaller communication radius r n According to the characteristics of the min-max problem, the larger the distance, the more adjustments are made to the direction. Therefore, ω n It is expressed as:
[0089]
[0090] For heterogeneous ground topology, the trajectory design method steps are as follows:
[0091] 1) Define n and l as the number of loops (the number of ground users and the number of iterations), and each ground user position g n , communication radius r n and the amount of data R n ; Define the initial position u0 and final position u of the drone F ;
[0092] 2) Define the step size λ, algorithm accuracy ε and maximum number of iterations I; initialize the starting waypoint β 1 =u0; initialize l = 2;
[0093] 3) for 1<n<N, calculate D according to formula (8) n n projection points on
[0094] 4) By solving the min-max problem in formula (12), the l+1th optimal solution β is obtained * :
[0095] 5) for 1<n<N, calculate the drone from β according to formula (5) l To β l+1 Total amount of data collected during the flight And calculate If R n <0, then delete g from the ground user N n ;
[0096] 6) l = l + 1, if l < I, return to step 3), otherwise, execute step 7);
[0097] 7) Return the waypoint set L.
[0098] Figure 5 Trajectory diagram of collecting data from six ground users for a UAV based on a homogeneous ground topology. The six ground users have the same communication boundaries and data volumes. Therefore, we set all segments r n =400m, step size parameter λ 1 =1,λ l =0.5, where l>1.
[0099] Figure 6 Trajectory diagram of collecting data from 6 ground users for a UAV based on a heterogeneous ground topology. N ground users have different communication radius and preset data transmission volume. The waypoints indicated by the blue dashed arrows and the waypoints represented by the ground users in blue fonts indicate that the UAV has completed the data collection task before the waypoints indicated by the arrows. Figure 6 It can be seen that since GU1 and GU2 need to transmit the largest amount of data, the first half of the trajectory is biased toward the left side of the ground topology as a whole.
[0100] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A method for joint optimization of data collection and trajectory planning based on drone assistance in wireless networks, characterized in that: Including the steps: (1) Constructing the system model: The UAV flies in a three-dimensional environment. The initial position of the UAV is at the origin (0, 0, 0), and the final position is (x F ,y F , H), further, assuming that the UAV flies horizontally at a fixed height H and a constant speed V, the data collected by n ground users is represented as N = {1, ..., N}, and the communication range of the nth ground user is represented as a radius r n Disc D n : 1) The time-varying distance from the nth ground user to the UAV is: Where u(t) = [x(t), y(t)] T represents the coordinates of the above position projected on the horizontal plane, g n =[a n , b n ] T Indicates the location of the nth ground user; 2) The instantaneous channel power gain from the nth ground user to the UAV is: where β0 represents the channel power gain at the reference distance d0 = 1m; B represents the total available channel bandwidth in hertz (Hz); 3) The instantaneous channel capacity from the UAV to the nth ground user is: where P is the transmit power from the ground user to the UAV, which is assumed to be constant during mission completion, and σ 2 is the additive white Gaussian noise (AWGN) of the nth terrestrial user uplink, γ0=β0P / σ 2 is the SNR received at d0 = 1m; 4) The constraint conditions are: Among them, the binary variable ξ n represents the communication execution period between the UAV and each ground user during the mission. If the UAV flies beyond the range D n If it is the inner area of , it is equal to 1, otherwise it is equal to 0; 5) The total amount of data sent from the nth to the UAV within the duration T is: (2) Based on parallel projection for waypoint positioning, we transform the trajectory design problem into a problem of the positions of a set of waypoints, which are connected in sequence by straight lines to form the UAV's trajectory. We will adjust the parallel projection method based on the ground topology to calculate the positions of the waypoints by redefining some rules of the parallel projection method; 1) The starting point x is in the convex set D n The only orthogonal projection onto is: 2) The single-step parallel projection function is: Among them, β l is the iteration point of PPM, λ l ∈[ε, 2-ε] where 0<ε<1 is called the relaxation parameter. Geometrically, it represents β l Move to Beta l+1 The parallel projection method is an improvement on the projection method in formula (6), which projects to all disks D in each iteration. n The centroid of the polygon formed by these N projection points is used as the next iteration point; 3) The improved parallel projection method is: Here, we assume that L = {1, ..., L} is the waypoint β on the trajectory of the UAV. l and β l =u0,β L =u F , β l =[x l ,y l ] T ; When input D n It will be used for each D in each MPPM iteration. n Projection to g n position, but not exceeding g n ; Leave D n After that, we only need to F In other cases, the current waypoint β l The closest D to it is projected in the traditional projection method n At the boundary point; 4) N projection points and waypoints β l+1 The min-max problem between is: Where β = [x, y] T is the optimization variable; 5) β l The iteration equation is: b l+1 =b l +λ l (b * -b l ),n∈N (10) From formula (9), we can get the optimal solution β * After that, we still use λ l To control β l The step length β * ; (3) Two trajectory design methods are respectively designed for homogeneous and heterogeneous ground topologies: For a homogeneous ground topology, each user has the same communication radius r n = r and unknown data collection volume; in this case, we only need to directly use the improved parallel projection method to calculate the waypoints of the trajectory; 1) Define n as the number of ground users, l as the number of cycles, and the position g of each ground user n , communication radius r n and the amount of data R n ; Define the initial position u0 and final position u of the drone F ; 2) Define the step size λ, algorithm accuracy ε and maximum number of iterations I; initialize the starting waypoint β l =u0; initialize l = 2; 3) For 1 < n < N, calculate D according to formula (8) in 1 n n projection points P on Dn (β l ); 4) By solving the min-max problem in formula (9), we obtain the l+1th optimal solution β * : b l+1 =b l +λ l (b * -b l ),n∈N 5) l = l + 1. If l < I is satisfied, return to step 3); otherwise, execute step 6); 6) Return the waypoint set L; For heterogeneous ground topology, each ground user may have a different communication radius r n and the preset amount of data collection; therefore, the UAV adds a new communication constraint during flight, namely, throughput, which can be expressed as: Where R n Indicates the amount of data for the nth item; To ensure communication fairness under throughput constraints, N projection points and waypoints β l+1 The min-max problem between can be refactored as: where ω n When the communication radius r is given n and the amount of data R n This makes the waypoints closer and the waypoints have a larger data volume R n and a smaller communication radius r n ; According to the characteristics of the minimum-maximum problem, the greater the distance, the more the UAV adjusts toward the ground user; therefore, ω n It is expressed as: For the heterogeneous ground topology, the steps of the trajectory design method are: 1) Define n and l as the number of loops, n is the number of ground users, l is the number of iterations, and each ground user position g n , communication radius r n and the amount of data R n ; Define the initial position u0 and final position u of the drone F ; 2) Define the step size λ, algorithm accuracy ε and maximum number of iterations I; initialize the starting waypoint β l =u0; initialize l = 2; 3) For 1 < n < N, calculate the n projection points P on Dn according to formula (8). Dn (β l ); 4) By solving the min-max problem in formula (12), obtain the (l + 1)th optimal solution β*: 5) For 1 < n < N, calculate the total amount of data collected by the UAV during the flight from β l to β l+1 and calculate that if R n < 0, then delete g from the ground users N n ; 6) l = l + 1. If l < I is satisfied, return to step 3); otherwise, execute step 7); 7) Return the waypoint set L.
2. A method for joint optimization of data collection and trajectory planning based on drone assistance in a wireless network according to claim 1, characterized in that: In a UAV-assisted wireless network system with multiple adjacent ground users, an optimal flight trajectory is designed for the data-collecting UAV for homogeneous and heterogeneous topologies on the premise of ensuring the transmission fairness of all ground users.
